Thermal studies of Co(II), Ni(II) and Cu(II) complexes of N,N′-bis(3,5-Di-t-butylsalicylidene)ethylenediamine

Article abstract of DOI:10.1007/s10973-008-8980-8

The thermal decomposition kinetics of sterically hindered salen type ligand (L) and its metal complexes [M=Co(II), Ni(II), Cu(II)] were investigated by thermogravimetric analysis. A direct insertion probe-mass spectrometer (DIP-MS) was used for the charac

Full text of DOI:10.1007/s10973-008-8980-8

Journal of Thermal Analysis and Calorimetry, Vol. 96 (2009) 1, 267–276
THERMAL STUDIES OF Co(II), Ni(II) AND Cu(II) COMPLEXES OF
N,N’-bis(3,5-Di-t-BUTYLSALICYLIDENE)ETHYLENEDIAMINE
1*
F. DoÈan , M. Ulusoy , Ö. F. Öztürk , . Kaya and B. Salih
2
1,3
1
3
1
†anakkale Onsekiz Mart University, Faculty of Arts and Science, Chemistry Department, †anakkale, Turkey
2Ege University, Faculty of Science, Chemistry Department, zmir, Turkey
3Hacettepe University, Faculty of Science, Chemistry Department, Ankara, Turkey
The thermal decomposition kinetics of sterically hindered salen type ligand (L) and its metal complexes [M=Co(II), Ni(II), Cu(II)]
were investigated by thermogravimetric analysis. A direct insertion probe-mass spectrometer (DIP-MS) was used for the
characterization of metal complexes of L and all fragmentations and stable ions were characterized. The thermogravimetry and
differential thermogravimetry (TG-DTG) plots of salen type salicylaldimine ligand and complexes showed a single step.
The kinetic analysis of thermogravimetric data was performed by using the invariant kinetic parameter method (IKP). The
values of the invariant activation energy, Einv and the invariant pre-exponential factor, Ainv, were calculated by using Coats–Red-
fern (CR) method. The thermal stabilities and activation energies of metal complexes of sterically hindered salen type ligand (L)
were found as Co(II)>Cu(II)>Ni(II)>L and ECu>ENi>ECo>L. Also, the probabilities of decomposition functions were investigated.
The diffusion functions (Dn) are most probable for the thermal decomposition of all complexes.
Keywords: activation energy, sterically hindered Schiff base, thermal decomposition kinetic, thermogravimetry
Introduction
trinuclear Schiff base complexes with acetate bridges
have been studied by Durmuê et al. [14]. Yeêilel [15],
Mohamed et al. [16] and DoÈan et al. [17] have in-
vestigated the thermal behavior, kinetic and ther-
modynamic parameters of the metal complexes and
salts with imidazole and benzimidazole in various
atmosphere using TG/DTG, DTA and elemental
analytical methods. In this paper, salen type ligand
and its Cu(II) [18], Ni(II) [19] and Co(II) complexes
were synthesized according to the published literature
and investigated their thermal properties. We use a
method developed by Lesnikovich et al. [20] at dif-
ferent heating rates to compute invariant kinetic para-
meters (IKP) of salen type ligand and its metal comp-
lexes, which have usually been applied to study the
decomposition of polymeric materials from thermo-
gravimetric data.
The coordination chemistry of salen type salicylaldimine
ligands with metal complexes have some interesting
applications, e.g. in catalytic oxidation reactions and
electrochemical reduction processes [1], in metallo-
enzyme-mediated catalysis [2], metal-containing liquid-
crystalline polymers [3] and as catalytically active
materials to develop surface-modified electrodes [4].
It is known that sterically hindered ligands bearing
salicylaldimines are effective antioxidant and widely
used in rancidification of fats and oils [5]. In recent
years, manuscripts dealing with thermal analysis of
salen type salicylaldimine ligands with metal complexes
are insufficient. The determination of kinetic parameter
from data of non-isothermal thermogravimetry is one
of the most difficult kinetic problems. Considerable
attention is paid to the kinetic parameters calculation
from TG curves. Many authors investigated the
kinetic data analysis of complexes by integral, differential Experimental
and specified methods. Also, new calculation methods
Preparation of samples
are still being published [6, 7]. A number of papers
are devoted to comparing these methods [8–11] or to
their critical assessment [12]. Soliman [13] has
investigated the thermal stabilities of the chromium and
molybdenum complexes with salicylidene-2-aminophenol,
salicylidene-2-aminoanisole, salicylidene-2-aminoaniline
and biquinoline. Thermal decomposition of some linear
Metal solutions of Co(II), Ni(II) and Cu(II) ions are
prepared from analytical purity reagents of
Co(CH3COO)2·4H2O, Ni(CH3COO)2·4H2O and
Cu(CH3COO)2·H2O salts. Metal complexes (CoL, CuL
and NiL) are obtained from the reactions of metal so-
lutions with alcoholic (ethanol) ligand solutions. The
*
Author for correspondence: fatihdogan@comu.edu.tr
1388–6150/$20.00
© 2009 Akadémiai Kiadó, Budapest
Akadémiai Kiadó, Budapest, Hungary
Springer, Dordrecht, The Netherlands

DO(AN et al.
analytical data and also FTIR results of CuL and NiL
atmosphere of nitrogen (10 cm3 min–1), sealed platinum
pan. All kinetic experiments were carried in duplicate
and reproducible results with the smaller variations in
the kinetic parameters were obtained. L and its metal
ion complexes, such as Cu(II), Ni(II) and
complexes are compatible with reported literature
[18, 19]. Solid metal complexes were filtered in vac-
uum and recrystallized from dichloromethane/ethanol
(v/v; 1/3). For Cu(II) complex: Yield: 85%; M.p.>270°C;
nmax (KBr Disk) (cm–1) 3033 (Ar-CH); 2985–2863
(Al-CH); 1615 (C=N); meff: 1.92 [B.M.]; lmax (nm)
(loge, M–1 L–1) in CHCl3: 300, 323 (4.19), 397 (4.11),
480a, 634 (2.42); Anal. found: C, 69.12; H, 7.88; N,
4.46. Calc. for C32H46N2O2Cu: C, 69.34; H, 8.37;
N, 5.05%. For Ni(II) complex: Yield: 87%; M.p.>270°C;
nmax (KBr Disk) (cm–1) 3028 (Ar-CH); 2992–2881
(Al-CH); 1620 (C=N); meff: diamagnetic; lmax (nm)
(loge, M–1 L–1) in CHCl3: 270 (4.37), 330 (3.98), 407
(3.80), 648 (2.19); Anal. found: C, 70.35; H, 8.39;
N, 5.33. Calc. for C32H46N2O2CuNi: C, 69.96; H,
8.44; N, 5.10%.
Co(II)-complexes, were analyzed using an Agilent 5973
Inert Mass Selective Detector equipped with Direct
Insertion Probe (HPP7&ProbeDirect, Scientific Instrument
Services, Ringoes, NJ USA). About 1 mg of the ligand
and its complexes were weighted and then inserted
inside the quartz sample tube (Scientific Instrument
Services, Ringoes, NJ USA) that was used for
inserting the sample inside mass spectrometer. At the
mass spectrometer, Electron Impact Ionization (EI)
source was used, the temperature of the source was set
to 140°C, vacuum was 1.5·10–6 Torr during the recording
of mass spectra. The Direct Insertion Probe program
was set as given below:
Initial temperature was set to 50°C for 10 min
and then temperature was increased up to 420°C with
10°C min–1 temperature ramp. Finally temperature
held at 420°C for 50 min.
Synthesis of Co(II) complex of
N,N’-bis(3,5-di-t-butylsalicylidene)-1,3-ethylenediamine
The 1 mmol Schiff base ligand dissolved in hot ethanol
(25 mL) and the appropriate 1 mmol Co(OAc)2·4H2O
was added under an argon atmosphere over a few
minutes causing the precipitation of the orange product.
The reaction was allowed to cool to room temperature
and stirred (3 h). The product was isolated by filtration
from the reaction mixture and dried under vacuum.
Recrystallized from dichloromethane/ethanol (v/v; 1/3).
Yield: 72%; M.p. 201–205°C; nmax (KBr Disk) (cm–1)
3043 (Ar-CH); 2953–2868 (Al-CH); 1620 (C=N);
meff: 4.05 [B.M.], S=3/2 in Td.; Anal. found: C, 70.67;
H, 8.95; N, 5.23. Calc. for C32H46N2O2Co: C, 69.92;
H, 8.44; N, 5.10%.
Results and discussion
Thermal behaviour and GC-MS analysis of Schiff
base and its metal complexes
The thermal behavior of L and its metal-complexes
[Co(II), Cu(II) and Ni(II)] at different heating rates 5,
10, 15, 20°C min–1 are presented in Figs 1–4, respectively.
As seen in Figs 1–4, each TG curves for both L and its
metal-complexes exhibits one stage decomposition.
The TG curve of all the compounds show a similar
characterizations. The thermal decomposition products,
the temperature range concerned (initial, Ti and final, Tf),
the percentage of mass lost during the decomposition,
the maximum decomposition temperature (Tm) and
the theoretical percentage of mass lost calculated
from TG curve at heating rate of 5°C min–1 are listed
in Table 1 for both L and its metal complexes.
Instrumentation
All thermal analysis work were performed on Perkin
Elmer Termogravimetric Analyzer Pyris-6, operating
in dynamic mode, with the following conditions; sample
mass ~5 mg, heating rate=5, 10, 15, 20°C min–1,
Table 1 Ti, Tm and Tf values of L, CoL, CuL and NiL for the heating rate of 5°C min–1
Mass loss/%,
DTGmax/°C
Complex
L
Step
Temperature range/°C
204–293
Assignment
C32H48N2O2
found
100
(calc.)
100
I
278
>293
401
residue
CoL
CuL
NiL
I
319–420
297–412
290–395
85.2
13.1
84.1
13.5
85.9
12.2
86.4
13.6
85.7
14.3
86.4
13.6
C32H46N2O2Co
CoO
residue
I
>420
395
C32H46N2O2Cu
CuO
resideu
I
>412
376
C32H46N2O2Ni
NiO
residue
>395
268
J. Therm. Anal. Cal., 96, 2009

THERMAL STUDIES OF Co(II), Ni(II) AND Cu(II) COMPLEXES
Fig. 2 TG/DTG curves of CuL at different rates
Fig. 1 TG/DTG curves of L at different rates
Fig. 4 TG/DTG curves of NiL at different rates
Fig. 3 TG/DTG curves of CoL at different rates
There is mainly one stage of decomposition of L
and its metal complexes for all the heating rates
investigated and a higher heating rate shifts the rate
curve to a higher temperature.
Also, in order to check the thermal stability of
the complexes in the solid state, initial temperatures
of the decomposition of L and its complexes were
compared. It was found that the thermal stabilities of
ligand and its metal complexes follow the order
Co(II)>Cu(II)>Ni(II)>L. In literature, it was found by
Arshad et al. that thermal stabilities of 1,2-dipi-
peridinoethane complexes of cobalt, nickel, zinc and
cadmium, increase in the following sequence: Ni(II)<
Co(II)<Cu(II)<Zn(II)<Cd(II) [21]. On the other hand,
in another study, the thermal stabilities of metal
complexes of the 6-(2-pyridylazo-acetamidophenol
were found as Ni(II)>Cu(II)>Zn(II)>Fe(II)>Co(II) [22].
Consequently, it is believed that the coordination of
the metal ions to ligands is responsible for thermal
stabilities of the metal complexes. The solid residue
of thermal decomposition of all the complexes was
metal oxide. The characterization of the end products
Fig. 5 X-ray powder diffraction pattern of CoO
of the decomposition was achieved by X-ray diffraction.
An as example the X-ray pattern of the end product of
Co(II) complexes is shown in Fig. 5. These results
conformed to our previous work [23].
Mass spectrum of ligand (L) is given in Fig. 6 with
the total ion chromatogram of L. On the total ion
chromatogram, beside the main peak of the ligand
obtained at 3.9 min retention time, some other deg-
radation peaks were observed. When the mass spectrum
of the main peak on total ion chromatogram was
evaluated, it was noticed that this compound pointed out
the mass spectrum of L exactly. In the mass spectrum
given in Fig. 6B, the peaks appeared at 492, 477, 260,
J. Therm. Anal. Cal., 96, 2009
269

DO(AN et al.
Table 2 Mass and intensity (%) of the peaks were evaluated from mass spectra of the studied complexes
Co(II) complex
Ni(II) complex
Cu(II) complex
Nominal Intensity Type of ion
Nominal mass Intensity Type of ion
Nominal
Intensity Type of ion
549
534
260
246
518
258
231
267
100
57
M+
548
533
259
550
535
260
57
100
M+
553
261
538
244
229
216
57
100
78.8
M+
[M-15]+
86.4 [M-15]+
82.6 [M-15-15]2+
44.1 [M+2]+
[M-15-15]2+
[M-15]+
57.6 [M-15-15]2+
9.4 [C16H24NO]+
5.0 [M-15-16]+
6.2 [C17H24NO]
3.1 [L-15-15]+
7.2
41.8
16.2
9.1
[C16H24NO]+
[C17H24NO-15-25]+
[C14H18NO]+
[(CH3)3C]+
38.5 [M-15+2]+
33.7 [M-15]+?
25.9 [(CH3)3C]+
8.0
8.5
[L-15-15]+
11.4
231
272
8.8
259, 244, 231, 218 and 57 atomic masses characterized
the M+, [M-15]+, [C17H26NO]+, [C17H25NO]+,
[C17H25NO-15]+, [M-15-15]2+, [C17H25NO-15-26]+ and
[(CH3)3C]+. Especially, the peak at 231 atomic mass
with high intensity is really interesting and representing
very stable doubly charged ion of the ligand in EI-mass
spectrometry which is unexpected. Doubly charged ion
stability is because of the high volume of ligand in the
space and the charged locations are enough far away
from each other to eliminate the same charged center
repulsions.
When mass spectra of metal complexes of this
ligand were interpreted, similar fragmentations were
observed but doubly charged ion intensity of the com-
plexes accompanied two methyl groups elimination
from both bulky sides of the complexes were found to
be higher than the intensity of the same peak of pure
ligand. For the metal complexes, the mass of the
peaks observed on mass spectra with their per cent in-
tensities are given in Table 2. In order to evaluate the
high intense doubly charged ion further, mass spectra
of nickel complex of L with the total ion mass
chromatogram of the complex is given in Fig. 7. Cho-
sen the nickel complex is the reason of the intense iso-
topic mass distribution of nickel in the complex. Eval-
uating the isotopic peak masses and intensities of
nickel in either single charge molecular ion or doubly
charged ion form, existing of doubly charged ion in
the spectrum could be examined clearly. Single
charged molecular ion at 548 and at 550 nominal
masses are the M+ and [M+2]+ characterizing the
highest abundant nickel isotopes which have 58 Ni
and 60 Ni with 68.08 and 26.22% naturally abun-
dance. The doubly charged ion of the complex formed
with leaving of two methyl groups from both bulky
sides of the complex yielded again two peaks ap-
peared at 260 and 259 nominal masses showing again
nickel isotope distribution
Fig. 6 A – Total ion chromatogram and B – mass spectrum of
ligand. Mass spectrum was obtained at 3.9 retention
time
Except 13C isotope addition. These results clearly
show that complex yields high intense (very stable) dou-
bly charged ion loosing two methyl groups.
Fig. 7 A – Total ion chromatogram and B – mass spectrum of
L-Ni(II) complex. Mass spectrum was obtained at 30.4
retention time
270
J. Therm. Anal. Cal., 96, 2009

THERMAL STUDIES OF Co(II), Ni(II) AND Cu(II) COMPLEXES
-E
Kinetic parameters
da
dT
A
= exp
æ
ç
ö
÷
a
f (a )
(3)
b
RT
è
ø
IKP method
The thermal decomposition of a solid in a heterogeneous
process is accompanied by the release of gaseous
products can usually be characterized by several forms
of the function f(a) given in Table 3. The invariant
kinetic parameter method is based on the relation of
the compensation effect which described by the
values of activation parameters Ea, and pre-exponential
factors A, obtained from various conversion function
using one of different methods (differential or integral
methods, etc.) for TG curves. To apply this method,
a=a(T) curves for several heating rates (bv, v=1,2,3)
are recorded and the activation parameters Ea and
pre-exponential factors A for a set of conversion function
(fj(a), j=1,2,3..), at each heating rates are calculated
using an integral, differential method and other method.
In this work, the CR method [26] was used to
integrate Eq. (3). It leads to the following relations:
Any kinetic analysis of solid state decomposition is
usually based on the rate equation
da
=kf (a )
(1)
dt
where a is the degree of conversion, t retention time
and f(a) the differential function of conversion. This
function may take various forms, some of which are
shown in Table 3 [24, 25].
The explicit temperature dependence of the rate
constant is introduced by replacing k=Aexp(–Ea/RT)
with the Arrhenius law, which gives
-E
da
æ
ö
÷
a
= Aexp
f (a )
(2)
ç
dt
RT
è
ø
where A is the pre-exponential factor (s–1), Ea the acti-
vation energy (J mol–1) and R the gas constant
(J mol–1 K–1). When the reaction is carried out by con-
trolling the temperature (T=T0+bt, where b is the lin-
ear heating rate and T0 the starting temperature)
Eq. (2) may be written as
E
gj (a iv
)
Ajv R
a
jv
ln
@ln
-
(4)
2
Tiv
bv E
RTiv
a
jv
with
Table 3 f(a) and g(a) functions used in the invariant kinetic parameter method
No.
Mechanisms
Symbol
Differential form, fj(a)
Integral form, gj(a)
1
2
N and G (n=1)
N and G (n=1.5)
N and G (n=2)
N and G (n=3)
N and G (n=4)
S1
S2
4(1–a)[–ln(1–a)]3/4
3(1–a)[–ln(1–a)]2/3
2(1–a)[–ln(1–a)]1/2
(3/2)(1–a)[–ln(1–a)]1/3
(1–a)
[–ln(1–a)]1/4
[–ln(1–a)]1/3
[–ln(1–a)]1/2
[–ln(1–a)]2/3
[–ln(1–a)]
a
3
S3
4
S4
5
S5
6
Contracted geometry shape (contracting linear)
Contracted geometry shape (cylindrical symmetry)
Contracted geometry shape (sphere symmetry)
Diffusion, 1D
S6
1
7
S7
2(1–a)1/2
2[1–(1–a)1/2]
3[1–(1–a)1/3]
8
S8
3(1–a)2/3
2
a
9
S9
1/(2a)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Diffusion, 2D
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
S20
S21
S22
S23
1/(ln(1–a))
(1–a)ln(1–a)+a
Diffusion, 3D
1.5/[(1–a)–1/3–1]
[1.5(1–a)2/3][1–(1–a)1/3]–1
(3/2)(1+a)2/3[(1+a) 1/3–1]–1
(3/2)(1–a)4/3[1/(1–a) 1/3–1]–1
(1–2a/3)–(1–a)2/3
2
]
Diffusion, 3D
[1–(1–a)1/3
Diffusion, 3D
[(1+a)1/3–1]2
Diffusion, 3D
[[1/(1–a)]1/3–1]2
3/4
4a
1/4
a
Mample power law
2/3
1/3
a
Mample power law (n=3)
Mample power law (n=2)
Mample power law (n=3/2)
Mample power law (n=2/3)
Mample power law (n=4/3)
Chemical reaction (n=2)
Chemical reaction (n=1/2)
Chemical reaction (n=1/3)
(1.5)a
1/2
2a
1/2
a
3/2(a)1/3
2/3(a)–1/2
4/3(a)–1/3
(1–a)2
2/3
a
3/2
a
3/4
a
(1–a)–1–1
1–(1–a)1/2
1–(1–a)1/3
2(1–a)1/2
3(1–a)2/3
J. Therm. Anal. Cal., 96, 2009
271

DO(AN et al.
Table 4 Apparent conversion parameters of L obtained for various conversation functions by means of IKP method
Heating rate
Function
5°C min–1
10°C min–1
15°C min–1
20°C min–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
S1
S2
18.890
28.110
46.540
64.970
101.83
76.510
88.000
92.340
161.77
175.80
181.60
193.43
148.52
233.64
12.565
19.670
33.881
48.090
119.13
55.196
92.340
88.000
136.81
0.860
3.280
7.820
12.19
20.72
14.37
17.27
18.35
33.29
36.05
35.97
38.86
27.76
48.65
–1.06
0.900
4.470
7.850
23.90
9.500
17.26
16.57
29.35
21.250
31.380
51.630
71.890
112.39
80.210
94.260
99.810
169.54
186.24
193.55
208.75
155.03
263.14
13.210
20.650
35.540
50.430
124.87
57.870
99.810
94.260
161.79
1.970
4.500
9.290
13.92
22.95
15.23
18.62
19.95
34.35
37.60
37.82
41.37
28.65
54.02
–0.34
1.620
5.230
8.640
24.86
10.31
18.85
17.93
34.56
16.910
25.640
43.110
60.570
95.500
69.120
80.880
85.410
147.54
161.69
167.72
180.11
134.81
223.25
10.310
16.850
29.920
42.980
108.33
49.520
85.410
80.880
133.83
1.090
3.320
7.460
11.41
19.10
12.77
15.61
16.70
29.23
31.87
31.77
34.66
23.97
44.65
–0.89
0.900
4.090
7.070
21.07
8.520
15.60
14.92
28.10
15.460
23.720
40.230
56.750
89.790
64.910
75.910
80.190
139.14
152.33
158.00
169.70
127.36
210.87
9.2400
15.430
27.800
40.170
102.02
46.350
80.190
75.910
126.66
0.930
3.050
6.950
10.68
17.89
11.95
14.59
15.61
27.33
29.73
29.55
32.27
22.30
41.78
–0.99
0.740
3.760
6.570
19.70
7.940
14.51
13.90
26.53
S3
S4
S5
S6
S7
S8
S9
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
S20
S21
S22
S23
a
da
g (a )=ò
j
f j (a )
0
where i is data point; b, v are the heating rate and its
numbers; j is the numbers of the conversion function be-
g (a
)
iv
é
ê
ë
ù
ú
û
1
j
tween 1 and 23. A plot ln
vs.
for a given
2
Tiv
Tiv
analytical form of g(a) should be a straight line. We use
the twenty three different conversion functions given in
Table 3 for the differential f(a) and integral g(a) func-
tions in Eq. (4). 23 couples (Eajv, Ajv) of L and its metal
complexes per heating rate, b are obtained by using the
CR method and are listed in Tables 4–7.
The application of the IKP method is based on
the study of the compensation effect. If a compensation
effect is observed, a linear relation defined by Eq. (5)
for each heating rate, is obtained. It is shown by plotting
lnA vs. Ea that a compensation effect is observed for
each heating rate (Figs 8 and 9). Thus, using the relation
of the compensation effect, for each heating rate the
compensation parameters (a*, b*) are determined.
Fig. 8 Compensation effect for the L and CuL vs. heating rates
lnA=a*+b*Ea
(5)
272
J. Therm. Anal. Cal., 96, 2009

THERMAL STUDIES OF Co(II), Ni(II) AND Cu(II) COMPLEXES
Table 5 Apparent conversion parameters of CuL obtained for various conversation functions by means of IKP method
Heating rate
Function
5°C min–1
10°C min–1
15°C min–1
20°C min–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
S1
S2
28.990
42.160
68.510
94.860
147.55
118.57
132.00
136.94
247.66
264.36
270.99
284.42
230.32
328.81
21.740
32.500
54.020
75.530
183.11
86.290
136.94
132.00
185.16
1.930
4.660
9.860
14.89
24.76
18.89
21.62
22.62
42.25
44.89
44.70
47.36
36.55
56.15
0.230
2.510
6.790
10.90
30.64
12.92
21.52
20.93
32.29
18.010
27.490
46.450
65.410
103.34
79.130
90.020
94.170
168.69
181.92
187.44
198.77
155.96
237.78
11.965
19.420
34.350
49.280
123.91
56.740
94.170
90.020
137.64
0.150
2.270
6.160
9.880
17.09
11.99
14.30
15.17
28.16
30.20
29.43
32.14
23.29
40.08
–1.45
0.310
3.450
6.380
20.15
7.800
14.07
13.60
24.18
15.800
24.590
42.170
59.760
94.930
82.800
88.650
90.690
176.20
183.90
186.46
191.97
166.35
209.31
12.760
20.550
36.110
51.670
129.49
59.460
90.690
88.650
108.80
–0.17
1.760
5.290
8.630
15.07
12.54
13.76
14.19
29.10
29.90
28.73
29.85
24.61
33.36
–0.98
0.780
3.940
6.880
20.75
8.320
13.09
13.07
17.95
18.050
27.670
46.910
66.150
104.64
77.910
89.740
94.340
166.64
180.87
186.94
199.51
153.54
243.70
11.370
18.760
33.550
48.340
122.28
55.730
94.340
89.740
144.20
0.690
2.770
6.620
10.28
17.37
11.92
14.35
15.29
26.90
29.15
29.20
30.10
22.56
40.43
–1.05
0.670
3.710
6.530
19.75
7.900
14.19
13.66
25.25
S3
S4
S5
S6
S7
S8
S9
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
S20
S21
S22
S23
which leads to the supercorrelation relation
a*=lnAint–b*Eaint
(6)
where a* and b* are constants (the compensation effect
parameters). A plot a* vs. b* is actually a straight line
whose parameters allow evaluation of the invariant acti-
vation parameters (Fig. 10). Thus, the values of invariant
activation energies and invariant pre-exponential factors can
be calculated from the slopes and intercepts of curves.
The values of a* and b* obtained for L and each
complexes are presented in Table 8.
The significance of a* and b* being characteristics
of the experimental conditions has been demonstrated
by Lesnikovich and Levchik [27].
Consequently, the values of the invariant activation
energy and the invariant pre-exponential factor calculated
from the slopes and intercepts of the straight lines
obtained by using Eq. (6) presented in Table 9.
Complexes of Cu, Ni and Co show similar
decomposition steps, the smaller size of Cu(II) as
compared to Co(II) and Ni(II) permits a closer approach of
the ligand to Cu(II) ion. Hence the Eint value for the
Cu(II) complex is higher than that of Co(II) and Ni(II).
The Eint values for thermal decomposition with respect
Fig. 9 Compensation effect for the NiL and CoL vs. heating rates
J. Therm. Anal. Cal., 96, 2009
273

DO(AN et al.
Table 6 Apparent conversion parameters of NiL obtained for various conversation functions by means of IKP method
Heating rate
Function
5°C min–1
10°C min–1
15°C min–1
20°C min–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
S1
S2
357.41
386.97
399.20
424.18
329.46
509.23
226.86
147.69
108.12
68.530
48.740
173.39
197.62
206.78
81.380
50.710
35.380
112.05
265.40
127.39
300.95
197.62
206.78
63.84
68.95
69.84
74.72
56.10
91.29
40.47
25.51
17.94
10.22
6.250
29.87
34.69
36.50
12.50
6.480
3.350
18.36
46.92
21.25
55.01
33.99
35.40
336.23
363.15
374.25
396.90
310.42
473.82
211.28
137.28
100.29
63.290
44.790
162.77
184.79
193.10
76.030
47.120
32.660
104.94
249.50
119.40
278.13
184.79
193.10
59.78
64.33
64.98
69.38
52.49
84.26
37.70
23.87
16.86
9.704
6.000
28.15
32.50
34.14
11.94
6.311
3.370
17.41
44.03
20.11
50.74
31.81
33.04
255.52
276.85
285.72
303.88
235.49
365.95
161.15
103.83
75.150
46.490
32.160
122.35
139.87
146.53
55.750
33.560
22.460
77.950
188.94
89.050
215.46
139.87
146.53
44.19
47.65
47.86
51.38
38.03
63.35
28.06
17.47
12.07
6.520
3.610
20.40
23.87
25.18
8.080
3.730
1.410
12.26
32.36
14.31
38.64
23.18
24.08
246.93
267.60
276.12
293.50
227.33
352.42
155.18
99.800
72.100
44.400
30.560
117.97
134.88
141.26
53.490
32.000
21.250
74.980
182.45
85.730
206.34
134.88
141.26
42.48
45.78
45.92
49.26
36.43
60.54
26.95
16.81
11.63
6.290
3.490
19.65
22.98
24.23
7.820
3.620
1.370
11.84
31.13
13.81
36.86
22.29
23.14
S3
S4
S5
S6
S7
S8
S9
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
S20
S21
S22
S23
Fig. 11 Probability distribution of the kinetic decomposition
functions of L computed with IKP method
Fig. 10 Verifying supercorrelation relation (Eq. (6))
experimental data with conversion function are de-
scribed by a complete and independent event system.
th
to the above method can be put into a descending order
as, ECu>ENi>ECo>L. The activation energies’ corres-
ponding decomposition steps are in the range of
66.68–180.17 kJ mol–1. On the other hand, the proba-
The probabilities of j function, Pj, developed by
Bourbigot [28, 29] for the decomposition process can be
2
S j
2
calculated by defining the ration Fj =
th
bilities of the j function, Pj, which may characterize the
using Ainv and
2
Sm
mode of decomposition of each sample, was discussion.
The Pj are computed with the assumption that the
274
J. Therm. Anal. Cal., 96, 2009

THERMAL STUDIES OF Co(II), Ni(II) AND Cu(II) COMPLEXES
Table 7 Apparent conversion parameters of CoL obtained for various conversation functions by means of IKP method
Heating rate
Function
5°C min–1
10°C min–1
15°C min–1
20°C min–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
Ea/kJ mol–1
lnA/s–1
S1
S2
41.900
59.460
94.570
129.69
199.92
159.75
178.46
185.30
330.26
353.62
362.81
381.37
305.76
442.34
31.860
46.070
74.490
102.91
245.00
117.12
185.30
178.46
251.19
4.660
8.160
14.94
21.57
34.62
26.71
30.41
31.75
57.63
61.48
59.76
65.36
50.62
77.15
2.460
5.350
10.86
16.21
42.24
18.86
30.66
29.71
44.63
44.460
62.920
99.840
136.77
210.61
156.86
180.80
190.03
324.64
353.47
365.74
390.99
298.41
479.01
31.020
45.010
72.970
100.93
240.75
114.92
190.03
180.80
288.89
5.770
9.400
16.44
23.32
36.90
26.50
31.15
32.93
56.55
61.38
62.22
67.03
49.27
83.74
2.880
5.700
11.07
16.28
41.59
18.86
31.84
30.45
51.88
44.970
63.660
101.06
138.45
213.24
160.98
184.50
193.47
333.08
361.63
373.59
398.07
306.49
482.12
31.900
46.240
74.930
103.61
247.03
117.95
193.47
184.50
287.06
6.090
9.700
16.70
23.55
37.04
27.11
31.59
33.30
57.37
62.05
62.79
67.37
50.11
83.06
3.330
6.170
11.57
16.82
42.30
19.41
32.20
30.90
50.93
37.700
54.030
86.100
119.36
184.70
127.73
153.47
163.25
266.76
298.05
311.11
337.79
239.03
428.76
23.460
35.040
58.210
81.390
197.25
92.980
163.65
153.47
264.19
4.910
8.120
14.29
20.30
32.13
21.31
26.22
28.08
45.47
50.61
51.53
56.85
38.02
73.73
1.830
4.210
8.670
12.95
33.38
15.06
26.80
25.96
46.72
S3
S4
S5
S6
S7
S8
S9
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
S20
S21
S22
S23
Table 8 The values of a* and b* for L, CuL, NiL and CoL vs. heating rates
L
CuL
b/°C min–1
a*
b*
r
a*
b*
r
5
10
15
20
–2.7482
–2.1152
–1.9223
–1.6495
0.2198
0.2130
0.2084
0.2036
0.99805
0.99810
0.99740
0.98995
–2.6420
–2.4496
–2.0882
–1.7482
0.1788
0.1783
0.1711
0.1708
0.99750
0.99700
0.99720
0.99190
NiL
CoL
b/°C min–1
a*
b*
r
a*
b*
r
5
10
15
20
–2.1926
–1.5878
–1.5647
–1.3357
0.1834
0.1810
0.1772
0.1754
0.99940
0.99935
0.99880
0.99870
–2.1731
–1.6896
–1.2825
–1.2307
0.1785
0.1781
0.1747
0.1742
0.99890
0.99925
0.99925
0.99890
r–regression coefficient
1 v = p
2
Eainv values obtained, where S j =
2
squares for each fi(a) and for each heating rate bv may
and Sm is the
åS
jv
be computed as
p i = 1
2
average minimum of residual dispersion. This ratio
th
obeys the F-distribution. Having n of the i of the
i =n
A
æ Einv
ö
÷fi (a jv ) (6)
da
dT
æ
ç
ö
-
÷
2
(n-1)S jv
=
inv expç
å
i =1
ç
÷
ø
bv
RTiv
è
øiv
è
experimental values of (da/dT)iv, the residual sum of
J. Therm. Anal. Cal., 96, 2009
275

DO(AN et al.
Table 9 Invariant kinetic parameters obtained for the salen
type salicylaldimine ligand and complexes at
different heating rate by means of CR method
References
1 W. Zhang, J. L. Loebach, S. R. Wilson and
E. N. Jacobsen, J. Am. Chem. Soc., 112 (1990) 2801.
2 T. Matsuura, Tetrahedron, 33 (1977) 2869.
3 J. L. Serrano and L. Oriol, Adv. Mater., 7 (1995) 365.
4 L. Mao, K. Yamamato, W. Zhou and L. Jin,
Electroanalysis, 12 (2000) 72.
Sample
Einv/kJ mol–1
Ainv/s–1
L
66.68
84.04
1.5·105
2.5·105
1.8·106
1.2·1013
CoL
NiL
CuL
89.74
5 J. F. Larrow, E. N. Jacobsen, Y. Gao, Y. Hong, X. Nie and
C. M. Zepp, J. Org. Chem., 59 (1994) 1939.
6 J. Madarász, G. Pokol and S. Gál, J. Thermal Anal.,
42 (1994) 539.
180.17
where n is the number of degrees of freedom equal for
every dispersion.
7 I. L. Lapides, J. Thermal Anal., 50 (1997) 269.
8 F. Carrasco, Thermochim. Acta, 213 (1993) 115.
9 M. V. Kök and N. Acar, J. Therm. Anal. Cal., 83 (2006) 445.
10 M. V. Kök, J. Therm. Anal. Cal., 88 (2007) 663.
11 M. V. Kök and A. G. Iscan, J. Therm. Anal. Cal.,
88 (2007) 657.
The kinetic pairs (Ainv and Eainv) and probabili-
ties for twe decomposition functions were carried out
with software developed in our laboratory using Ex-
cel 2000. The histogram showing the distribution of the
probabilities associated with the twenty-three decom-
position functions is presented for L in Fig. 11.
The kinetic functions of the four materials are
close. From the distribution of probabilities can be
shown that the diffusion functions (Dn) are most
probable for the thermal decomposition of each
sample. The histograms obtained from differential
methods are also similar that of integral method.
12 J. Zsakó, J. Thermal Anal., 46 (1996) 1854.
13 A. A. Soliman, S. M. El-Medani and O. A. M. Ali,
J. Therm. Anal. Cal., 83 (2006) 385.
14 S. Durmuê, Ü. Ergün, J. C. Jaud, K. C Emregül, H. Fues
and O Atakol, J. Therm. Anal. Cal., 86 (2006) 337.
15 O. Z. Yeêilel, H. Ölmez and H. ¸budak, J. Therm. Anal.
Cal., OnlineFirst DOI: 10. 1007/s10973-005-7479-9.
16 G. G. Mohamed and H. Z. El-Wahab, J. Therm. Anal.
Cal., 73 (2003) 347.
17 F. DoÈan, S. Gülcemal, M. Yürekli and B. †etinkaya,
J. Therm. Anal. Cal., 91 (2008) 395.
18 V. T. Kasumov and F. Koksal, Spectrochim. Acta, Part A,
61A(1-2) (2004) 225.
Conclusions
19 V. T. Kasumov, S. O.-Yaman and E. Tas, Spectrochim.
Acta, Part A, 62A(1-3) (2005) 716.
The TG curves of all the compounds show similar de-
composition when heated to about 500°C. The com-
plexes prepared with different metals decompose via
a one-step process. It was found that the thermal sta-
bilities of the ligand and its metal complexes follow
the order Co(II)>Cu(II)>Ni(II)>L. Using the IKP
method a kinetic analysis of the thermogravimetric data
was performed. The activation energy of the com-
plexes determined by applying Coats and Redfern
method and increased in the order Co(II)<Ni(II)<Cu(II).
Also, the activation energy of L and the complex sam-
ples studied was in the range of 66–180 kJ mol–1 de-
pending on the heating rate and model assumptions.
The kinetic function related to the mode of decomposition
of all complexes was found to be close. The diffusion
functions (Dn) are most probable for the thermal de-
composition of all complexes. DIP-MS was found to
be very powerful technique for the characterization of
the complex structure. It was noticed that doubly
charged ion stability occurred via two methyl groups
leaving from both sides of the ligand and its metal ion
complexes was high.
20 S. V Levchik, G. F. Levchik and A. L. Lesnikovich,
Thermochim. Acta, 92 (1985) 157.
21 M. Arshad, Saeed-ur-Rehman, S. Ali Khan, K. Masud,
N. Arshad and A. Ghani, Thermochim. Acta, 364 (2002) 143.
22 G. G. Mohamed, F. A. Nour-El Dien and E. A. El-Gamel,
J. Therm. Anal. Cal., 67 (2002) 135.
23 F. DoÈan, M. Ulusoy, Ö. F. Öztürk, I. Kaya and B. Salih,
J. Therm. Anal. Cal., (2009), in print.
24 V. Mamleev, S. Bourbigot, M. Le Bras, S. Duquesne and
J. Seëták, Phys. Chem. Chem. Phys., 2 (2000) 4708.
25 A. J. Lesnickovick and S. V Levchick, J. Thermal Anal.,
30 (1985) 667.
26 A. W. Coats and J. P. Redfern, Nature, 201 (1964) 68.
27 A. J. Lesnickovick and S. V Levchick, J. Thermal Anal.,
27 (1983) 83.
28 S. Bourbigot, R. Delobel, M. Le Bras and D. Normand,
J. Chim. Phys., 90 (1993) 1909.
29 S. Bourbigot, X. Flambard and S. Duquesne, Polym. Int.,
50 (2001) 157.
Received: January 7, 2008
Accepted: September 16, 2008
DOI: 10.1007/s10973-008-8980-8
Acknowledgements
The authors gratefully acknowledge Basri Gülbakan and Ömür
†elikbi¸ak for helpful Mass Spectrometry measurements.
276
J. Therm. Anal. Cal., 96, 2009